A New Type of Model Validity Used in Linear Combining Forecasting Model DOI Creative Commons

Zhao Zhang

Journal of Mathematics, Journal Year: 2024, Volume and Issue: 2024(1)

Published: Jan. 1, 2024

When combining numerous individual forecasting models linearly, it is common to construct a measure of model validity based on AD() or APE(). Specifically, we generally by using the mean and standard deviation AD APE. The resulting from this approach will serve as basis for assigning weights, then weights be used combine multiple into linear (LCFM).However, drawback APE alone that they do not consider varying importance between recent errors long‐term errors. To address limitation, article reconstructs new type (New Model Validity). sliding average fitting accuracy are two components New Validity, enhances incorporating smoothing coefficient. Thus when applying Validity LCFM, assigned higher weight if its more accurate in period. Through proof, shown long appropriate obtained, LCFM remains superior each under Validity. Therefore, feasible attempt improve performance

Language: Английский

The development trend of China’s marine economy: a predictive analysis based on industry level DOI Creative Commons
Yu Chen,

Huahan Zhang,

Lingling Pei

et al.

Frontiers in Marine Science, Journal Year: 2025, Volume and Issue: 12

Published: Feb. 10, 2025

This paper aims to provide insights into the future trends for marine industries in China, by forecasting added value key sectors and then offering tailored policy recommendations. Those economic indicators at industry level are characterized small sample sizes, sectoral heterogeneity, irregular fluctuations, which require a specialized methodology handle data features predictions each industry. To address these issues, conformable fractional grey model ( CFGM ), integrates accumulation with model, is applied proven effective through accuracy robustness tests. First, results from multi-step experiments demonstrate that significantly outperforms traditional statistical, machine learning models, models context of predictions, an average improvement 32.14%. Second, stability predictive values generated further verified Probability Density Analysis PDA ) multiple comparisons best MCB tests, thereby ruling out possibility accurate result mere chance. Third, used estimate across industries, accompanied suggestions ensure sustainable development economy.

Language: Английский

Citations

0
A New Type of Model Validity Used in Linear Combining Forecasting Model DOI Creative Commons

Zhao Zhang

Journal of Mathematics, Journal Year: 2024, Volume and Issue: 2024(1)

Published: Jan. 1, 2024

When combining numerous individual forecasting models linearly, it is common to construct a measure of model validity based on AD() or APE(). Specifically, we generally by using the mean and standard deviation AD APE. The resulting from this approach will serve as basis for assigning weights, then weights be used combine multiple into linear (LCFM).However, drawback APE alone that they do not consider varying importance between recent errors long‐term errors. To address limitation, article reconstructs new type (New Model Validity). sliding average fitting accuracy are two components New Validity, enhances incorporating smoothing coefficient. Thus when applying Validity LCFM, assigned higher weight if its more accurate in period. Through proof, shown long appropriate obtained, LCFM remains superior each under Validity. Therefore, feasible attempt improve performance

Language: Английский

Citations

0